Abstract:A preference matrix $M$ has an entry for each pair of candidates in an election whose value $p_{ij}$ represents the proportion of voters that prefer candidate $i$ over candidate $j$. The matrix is rationalizable if it is consistent with a set of voters whose preferences are total orders. A celebrated open problem asks for a concise characterization of rationalizable preference matrices. In this paper, we generalize this matrix rationalizability question and study when a preference matrix is consistent with a set of voters whose preferences are partial orders of width $\alpha$. The width (the maximum cardinality of an antichain) of the partial order is a natural measure of the rationality of a voter; indeed, a partial order of width $1$ is a total order. Our primary focus concerns the rationality number, the minimum width required to rationalize a preference matrix. We present two main results. The first concerns the class of half-integral preference matrices, where we show the key parameter required in evaluating the rationality number is the chromatic number of the undirected unanimity graph associated with the preference matrix $M$. The second concerns the class of integral preference matrices, where we show the key parameter now is the dichromatic number of the directed voting graph associated with $M$.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is about the rationalization of preference matrices. Specifically, it is about how to determine the measure of rationality when the preference matrix is consistent with a set of voters with partial orders. Traditionally, the rationalization problem of preference matrices is to study whether a set of voters with total orders can be found so that the preference matrix conforms to the preferences of these voters. However, this problem is very complex and difficult to solve. Therefore, the author of this paper generalizes the problem to partial orders and studies the situation when the preference matrix is consistent with a set of voters with partial orders of width α. The width refers to the maximum cardinality of the antichain in the partial order, which naturally measures the rationality degree of voters. A partial order with a width of 1 is a total order. The main focus of the article is on the **rationality number**, that is, the minimum width α such that the preference matrix can be interpreted as the preference result of a set of voters with partial orders whose width does not exceed α. ### Main Contributions 1. **Half - Integral Preference Matrices**: - For half - integral preference matrices, the author proves that the key parameter for evaluating the rationality number is the chromatic number of the unanimity graph related to the preference matrix. Specifically, the rationality number α(M) is bounded by the chromatic number χ(GM) of the unanimity graph. - Theorem: For the half - integral preference matrix class M1/2, we have \[ \frac{1}{5}\chi(G_M)\leq \exists \alpha(M)\leq \forall \chi(G_M) \] 2. **Integral Preference Matrices**: - For integral preference matrices, the author proves that the key parameter for evaluating the rationality number is the dichromatic number of the voting graph. This gives more precise boundaries for the rationality number. - Theorem: For the integral preference matrix class M0/1, we have \[ \frac{\chi(G_M)}{2\log n + 1}\leq \exists \alpha(M)\leq \forall \frac{3\chi(G_M)}{\log n} \] ### Example Illustrations - **Example I**: An unrationalizable integral preference matrix can be explained by a voter with a partial order of width 2. - **Example II**: An unrationalizable half - integral preference matrix requires at least two voters with partial orders of width 2 to be explained. - **Example III**: A general preference matrix can be explained by a voter where all candidates are not strict, and the rationality number is n at this time. ### Conclusion By introducing the concepts of partial orders and rationality numbers, this paper provides a new method to evaluate the rationalization problem of preference matrices and gives specific theoretical results in different types of preference matrices.

Matrix Rationalization via Partial Orders

Generalization of real interval matrices to other fields

Novel Theory for Polynomial and Rational Matrices at Infinity. Part 2. Rational Matrices

Interval matrices: realization of ranks by rational matrices

Voting with Partial Orders: The Plurality and Anti-Plurality Classes

Bounds on the moduli of eigenvalues of rational matrices

Maximum Stable Matching with Matroids and Partial Orders

Optimal $(0,1)$-Matrix Completion with Majorization Ordered Objectives (To the memory of Pravin Varaiya)

The Voter Basis and the Admissibility of Tree Characters

Orders with few rational monogenizations

Fractional Minimal Rank

On Locally Rationalizable Social Choice Functions

Unitary rational functions: The scaled quaternion case

Novel Theory for Polynomial and Rational Matrices at Infinity. Part1. Polynomial Matrices

Group Theoretical Characterizations of Rationality

Rational exponents for cliques

Statistics of ranks, determinants and characteristic polynomials of rational matrices

Dyadic linear programming and extensions

Partial Ordering of Hyper-powersets and Matrix Representation of Belief Functions within DSmT

Partial representations of orderings

Para-Hermitian rational matrices